RESUMEN

Alpha-helical transmembrane proteins (αTMPs) play essential roles in drug targeting and disease treatments. Due to the challenges of using experimental methods to determine their structure, αTMPs have far fewer known structures than soluble proteins. The topology of transmembrane proteins (TMPs) can determine the spatial conformation relative to the membrane, while the secondary structure helps to identify their functional domain. They are highly correlated on αTMPs sequences, and achieving a merge prediction is instructive for further understanding the structure and function of αTMPs. In this study, we implemented a hybrid model combining Deep Learning Neural Networks (DNNs) with a Class Hidden Markov Model (CHMM), namely HDNNtopss. DNNs extract rich contextual features through stacked attention-enhanced Bidirectional Long Short-Term Memory (BiLSTM) networks and Convolutional Neural Networks (CNNs), and CHMM captures state-associative temporal features. The hybrid model not only reasonably considers the probability of the state path but also has a fitting and feature-extraction capability for deep learning, which enables flexible prediction and makes the resulting sequence more biologically meaningful. It outperforms current advanced merge-prediction methods with a Q4 of 0.779 and an MCC of 0.673 on the independent test dataset, which have practical, solid significance. In comparison to advanced prediction methods for topological and secondary structures, it achieves the highest topology prediction with a Q2 of 0.884, which has a strong comprehensive performance. At the same time, we implemented a joint training method, Co-HDNNtopss, and achieved a good performance to provide an important reference for similar hybrid-model training.

Asunto(s)

RESUMEN

The conservation of avian diversity plays a critical role in maintaining ecological balance and ecosystem function, as well as having a profound impact on human survival and livelihood. With species' continuous and rapid decline, information and intelligent technology have provided innovative knowledge about how functional biological diversity interacts with environmental changes. Especially in complex natural scenes, identifying bird species with a real-time and accurate pattern is vital to protect the ecological environment and maintain biodiversity changes. Aiming at the fine-grained problem in bird image recognition, this paper proposes a fine-grained detection neural network based on optimizing the YOLOV5 structure via a graph pyramid attention convolution operation. Firstly, the Cross Stage Partial (CSP) structure is introduced to a brand-new backbone classification network (GPA-Net) for significantly reducing the whole model's parameters. Then, the graph pyramid structure is applied to learn the bird image features of different scales, which enhances the fine-grained learning ability and embeds high-order features to reduce parameters. Thirdly, YOLOV5 with the soft non-maximum suppression (NMS) strategy is adopted to design the detector composition, improving the detection capability for small targets. Detailed experiments demonstrated that the proposed model achieves better or equivalent accuracy results, over-performing current advanced models in bird species identification, and is more stable and suitable for practical applications in biodiversity conservation.

Asunto(s)

RESUMEN

INTRODUCTION: Sarcopenia is a muscle disease that involves loss of muscle strength and physical function and is associated with adverse health effects. Even though sarcopenia has attracted increasing attention in the literature, many research findings have not yet been translated into clinical practice. In this article, we aim to validate a deep learning neural network for automated segmentation of L3 CT slices and aim to explore the potential for clinical utilization of such a tool for clinical practice. MATERIALS AND METHODS: A deep learning neural network was trained on a multi-centre collection of 3413 abdominal cancer surgery subjects to automatically segment muscle, subcutaneous and visceral adipose tissue at the L3 lumbar vertebral level. 536 Polytrauma subjects were used as an independent test set to show generalizability. The Dice Similarity Coefficient was calculated to validate the geometric similarity. Quantitative agreement was quantified using Bland-Altman's Limits of Agreement interval and Lin's Concordance Correlation Coefficient. To determine the potential clinical usability, randomly selected segmentation images were presented to a panel of experienced clinicians to rate on a Likert scale. RESULTS: Deep learning results gave excellent agreement versus a human expert operator for all of the body composition indices, with Concordance Correlation Coefficient for skeletal muscle index of 0.92, Skeletal muscle radiation attenuation 0.94, Visceral Adipose Tissue index 0.99 and Subcutaneous Adipose Tissue Index 0.99. Triple-blinded visual assessment of segmentation by clinicians correlated only to the Dice coefficient, but had no association to quantitative body composition metrics which were accurate irrespective of clinicians' visual rating. CONCLUSION: A deep learning method for automatic segmentation of truncal muscle, visceral and subcutaneous adipose tissue on individual L3 CT slices has been independently validated against expert human-generated results for an enlarged polytrauma registry dataset. Time efficiency, consistency and high accuracy relative to human experts suggest that quantitative body composition analysis with deep learning should is a promising tool for clinical application in a hospital setting.

Asunto(s)

RESUMEN

The monthly high-resolution terrestrial water storage anomalies (TWSA) during the 11-months of gap between GRACE (Gravity Recovery And Climate Experiment) and its successor GRACE-FO (-Follow On) missions are missing. The continuity of the GRACE-like TWSA series with commensurate accuracy is of great importance for the improvement of hydrologic models both at global and regional scales. While previous efforts to bridge this gap, though without achieving GRACE-like spatial resolutions and/or accuracy have been performed, high-quality TWSA simulations at global scale are still lacking. Here, we use a suite of deep learning (DL) architectures, convolutional neural networks (CNN), deep convolutional autoencoders (DCAE), and Bayesian convolutional neural networks (BCNN), with training datasets including GRACE/-FO mascon and Swarm gravimetry, ECMWF Reanalysis-5 data, normalized time tag information to reconstruct global land TWSA maps, at a much higher resolution (100 km full wavelength) than that of GRACE/-FO, and effectively bridge the 11-month data gap globally. Contrary to previous studies, we applied no prior de-trending or de-seasoning to avoid biasing/aliasing the simulations induced by interannual or longer climate signals and extreme weather episodes. We show the contribution of Swarm and time inputs which significantly improved the TWSA simulations in particular for correct prediction of the trend component. Our results also show that external validation with independent data when filling large data gaps within spatio-temporal time series of geophysical signals is mandatory to maintain the robustness of the simulation results. The results and comparisons with previous studies and the adopted DL methods demonstrate the superior performance of DCAE. Validations of our DCAE-based TWSA simulations with independent datasets, including in situ groundwater level, Interferometric Synthetic Aperture Radar measured land subsidence rate (e.g. Central Valley), occurrence/timing of severe flash flood (e.g. South Asian Floods) and drought (e.g. Northern Great Plain, North America) events occurred within the gap, reveal excellent agreements.

Asunto(s)

RESUMEN

Training deep learning based handwritten text recognition systems needs a lot of data in terms of text images and their corresponding annotations. One way to deal with this issue is to use data augmentation techniques to increase the amount of training data. Generative Adversarial Networks (GANs) based data augmentation techniques are popular in literature especially in tasks related to images. However, specific challenges need to be addressed in order to effectively use GANs for data augmentation in the domain of text recognition. Text data is inherently imbalanced in terms of frequency of different characters appearing in training samples and the training data as a whole. GANs trained on the imbalanced dataset leads to augmented data that does not represent the minority characters well. In this paper, we present an adaptive data augmentation technique using GANs that deals with the issue of class imbalance arising in text recognition problems. We show, using experimental evaluations on two publicly available datasets for handwritten Arabic text recognition, that the GANs trained using the presented technique is effective in dealing with class imbalanced problem by generating augmented data that is balanced in terms of character frequencies. The resulting text recognition systems trained on the balanced augmented data improves the text recognition accuracy as compared to the systems trained using standard techniques.

RESUMEN

Left atrial (LA) enlargement and dysfunction increase the risk of atrial fibrillation (AF). Traditional echocardiographic evaluation of the left atrium has been limited to dimensional and semi-quantification measurement of the atrial component of ventricular filling, with routine measurement of LA function not yet implemented. However, functional parameters, such as LA emptying fraction (LAEF), may be more sensitive markers for detecting AF-related changes than LA enlargement. Speckle-tracking echocardiography has proven to be a feasible and reproducible technology for the direct evaluation of LA function. The clinical application, advantages, and limitations of LA strain and strain rate need to be fully understood. Furthermore, the prognostic value and utility of this technique in making therapeutic decisions for patients with AF need further elucidation. Deep learning neural networks have been successfully adapted to specific tasks in echocardiographic image analysis, and fully automated measurements based on artificial intelligence could facilitate the clinical diagnostic use of LA speckle-tracking images for classification of AF ablation outcome. This review describes the fundamental concepts and a brief overview of the prognostic utility of LA size, LAEF, LA strain and strain rate analyses, and the clinical implications of the use of these measures.

RESUMEN

In this study, a deep learning bidirectional long short-term memory (BiLSTM) recurrent neural network-based channel state information estimator is proposed for 5G orthogonal frequency-division multiplexing systems. The proposed estimator is a pilot-dependent estimator and follows the online learning approach in the training phase and the offline approach in the practical implementation phase. The estimator does not deal with complete a priori certainty for channels' statistics and attains superior performance in the presence of a limited number of pilots. A comparative study is conducted using three classification layers that use loss functions: mean absolute error, cross entropy function for kth mutually exclusive classes and sum of squared of the errors. The Adam, RMSProp, SGdm, and Adadelat optimisation algorithms are used to evaluate the performance of the proposed estimator using each classification layer. In terms of symbol error rate and accuracy metrics, the proposed estimator outperforms long short-term memory (LSTM) neural network-based channel state information, least squares and minimum mean square error estimators under different simulation conditions. The computational and training time complexities for deep learning BiLSTM- and LSTM-based estimators are provided. Given that the proposed estimator relies on the deep learning neural network approach, where it can analyse massive data, recognise statistical dependencies and characteristics, develop relationships between features and generalise the accrued knowledge for new datasets that it has not seen before, the approach is promising for any 5G and beyond communication system.

RESUMEN

A reliable assessment of the aquifer contamination vulnerability is essential for the conservation and management of groundwater resources. In this study, a recent technique in artificial intelligence modeling and computational optimization algorithms have been adopted to enhance the groundwater contamination vulnerability assessment. The original DRASTIC model (ODM) suffers from the inherited subjectivity and a lack of robustness to assess the final aquifer vulnerability to nitrate contamination. To overcome the drawbacks of the ODM, and to maximize the accuracy of the final contamination vulnerability index, two levels of modeling strategy were proposed. The first modeling strategy used particle swarm optimization (PSO) and differential evolution (DE) algorithms to determine the effective weights of DRASTIC parameters and to produce new indices of ODVI-PSO and ODVI-DE based on the ODM formula. For strategy-2, a deep learning neural networks (DLNN) model used two indices resulting from strategy-1 as the input data. The adjusted vulnerability index in strategy-2 using the DLNN model showed more superior performance compared to the other index models when it was validated for nitrate values. Study results affirmed the capability of the DLNN model in strategy-2 to extract the further information from ODVI-PSO and ODVI-DE indices. This research concluded that strategy-2 provided higher accuracy for modeling the aquifer contamination vulnerability in the study area and established the efficient applicability for the aquifer contamination vulnerability modeling.

Asunto(s)

RESUMEN

Hemorrhagic stroke is a medical emergency. Artificial intelligence techniques and algorithms may be used to automatically detect and quantitate intracranial hemorrhage in a semiautomated fashion. This article reviews the use of deep learning convolutional neural networks for managing hemorrhagic stroke. Such a capability may be used to alert appropriate care teams, make decisions about patient transport from a primary care center to a comprehensive stroke center, and assist in treatment selection. This article reviews artificial intelligence algorithms for intracranial hemorrhage detection, quantification, and prognostication. Multiple algorithms currently being explored are described and illustrated with the help of examples.

Asunto(s)

RESUMEN

Acute ischemic stroke constitutes approximately 85% of strokes. Most strokes occur in community settings; thus, automatic algorithms techniques are attractive for managing these cases. This article reviews the use of deep learning convolutional neural networks in the management of ischemic stroke. Artificial intelligence-based algorithms may be used in patient triage to detect and sound the alarm based on early imaging, alert care teams, and assist in treatment selection. This article reviews algorithms for artificial intelligence techniques that may be used to detect and localize acute ischemic stroke. We describe artificial intelligence algorithms for these tasks and illustrate them with examples.

Asunto(s)

RESUMEN

BACKGROUND/AIM: Mathematical models have long been considered as important tools in cancer biology and therapy. Herein, we present an advanced non-linear mathematical model that can predict accurately the effect of an anticancer agent on the growth of a solid tumor. MATERIALS AND METHODS: Advanced non-linear mathematical optimization techniques and human-to-mouse experimental data were used to develop a tumor growth inhibition (TGI) estimation model. RESULTS: Using this mathematical model, we could accurately predict the tumor mass in a human-to-mouse pancreatic ductal adenocarcinoma (PDAC) xenograft under gemcitabine treatment up to five time periods (points) ahead of the last treatment. CONCLUSION: The ability of the identified TGI dynamic model to perform satisfactory short-term predictions of the tumor growth for up to five time periods ahead was investigated, evaluated and validated for the first time. Such a prediction model could not only assist the pre-clinical testing of putative anticancer agents, but also the early modification of a chemotherapy schedule towards increased efficacy.

Asunto(s)

RESUMEN

One of the major challenges for protein docking methods is to accurately discriminate native-like structures from false positives. Docking methods are often inaccurate and the results have to be refined and re-ranked to obtain native-like complexes and remove outliers. In a previous work, we introduced AccuRefiner, a machine learning based tool for refining protein-protein complexes. Given a docked complex, the refinement tool produces a small set of refined versions of the input complex, with lower root-mean-square-deviation (RMSD) of atomic positions with respect to the native structure. The method employs a unique ranking tool that accurately predicts the RMSD of docked complexes with respect to the native structure. In this work, we use a deep learning network with a similar set of features and five layers. We show that a properly trained deep learning network can accurately predict the RMSD of a docked complex with 1.40 Å error margin on average, by approximating the complex relationship between a wide set of scoring function terms and the RMSD of a docked structure. The network was trained on 35000 unbound docking complexes generated by RosettaDock. We tested our method on 25 different putative docked complexes produced also by RosettaDock for five proteins that were not included in the training data. The results demonstrate that the high accuracy of the ranking tool enables AccuRefiner to consistently choose the refinement candidates with lower RMSD values compared to the coarsely docked input structures.

Asunto(s)